Background: The microbial production of biofuels is complicated by a tradeoff between yield and toxicity of many fuels. Efflux pumps enable bacteria to tolerate toxic substances by their removal from the cells while bypassing the periplasm. Their use for the microbial production of biofuels can help to improve cell survival, product recovery, and productivity. However, no native efflux pump is known to act on the class of short ‑ chain alcohols, important next‑ generation biofuels, and it was considered unlikely that such an efflux pump exists. Results: We report that controlled expression of the RND‑ type efflux pump TtgABC from Pseudomonas putida DOT‑ T1E strongly improved cell survival in highly toxic levels of the next‑ generation biofuels n‑ butanol, isobutanol, isoprenol, and isopentanol. GC‑ FID measurements indicated active efflux of n‑ butanol when the pump is expressed. Conversely, pump expression did not lead to faster growth in media supplemented with low concentrations of n‑ butanol and isopentanol. Conclusions: TtgABC is the first native efflux pump shown to act on multiple short ‑ chain alcohols. Its controlled expression can be used to improve cell survival and increase production of biofuels as an orthogonal approach to metabolic engineering. Together with the increased interest in P. putida for metabolic engineering due to its flexible metabolism, high native tolerance to toxic substances, and various applications of engineering its metabolism, our findings endorse the strain as an excellent biocatalyst for the high ‑ yield production of next‑ generation biofuels. Keywords: Higher alcohols, Short‑ chain alcohols, Next‑ generation biofuels, Efflux pumps, TtgABC, Pseudomonas putida, Tolerance, Toxicity microbial production, particularly of toxic substances, Background such as antibiotics or biofuels [5–7]. The high tolerance of several gram-negative bacte - Various strains of Pseudomonas are known for their ria to toxic substances compared to other organisms ability to metabolize a number of industrial products is attributed to a lower outer membrane permeability, and solvents as sole carbon source [8, 9], and to toler- periplasmic and cytosolic enzymatic degradation (e.g., ate high concentrations of toxic aromatic compounds β-lactamases), homeoviscous membrane adaptation, and [10–13]. The strain Pseudomonas putida DOT-T1E was multidrug efflux pumps [ 1–3]. The latter have the poten - isolated from a wastewater plant in Spain , and pos- tial not only to improve cell survival in toxic environ- sesses a number of efflux systems. TtgABC is an efflux ments, but also to directly increase yield and productivity system of the resistance-nodulation-cell division (RND) of production strains by removing the final product from type consisting of the inner membrane protein TtgB, the the cells and facilitating extracellular product recovery membrane fusion protein TtgA, and the outer membrane . Consequently, the discovery of efflux pumps act - channel TtgC [15, 16]. TtgABC is expressed constitutively ing on a target product of interest can help to improve , but is also assumed to be induced by n-butanol . It acts on specific aromatic hydrocarbons, such as *Correspondence: basler@mpimp‑golm.mpg.de toluene, m-xylene, and 1,2,4-trichlorobenzene, as well as Department of Chemical and Biomolecular Engineering, University antibiotics , in combination with the efflux systems of California, Berkeley, CA, USA Full list of author information is available at the end of the article © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Basler et al. Biotechnol Biofuels (2018) 11:136 Page 2 of 10 TtgDEF and TtgGHI . While the action of these and l-arabinose inducible promoter P on a broad host BAD other RND-type efflux systems on antibiotics and aro - range plasmid (BBR1) and transformed it into P. putida matic compounds is well-studied, their effect on short- DOT-T1E (“Methods”). The plasmid remained stable in P. chain alcohols, particularly relevant as next-generation putida for about 2 days without selective antibiotic while biofuels, is thus far not known. expressing TtgABC (Additional file 1: Figure S1). In E. Moreover, a large-scale screen for identification of coli, induction via the P promoter was shown to result BAD bacterial efflux pumps acting on biofuels returned no in non-homogenous expression levels, whereby different candidates for the short-chain alcohols n-butanol and sub-populations exhibit either high or no expression at isobutanol . Together with several other attempts intermediate inducer concentrations . Since this phe- to identify efflux pumps for biofuels [19, 20], this led to nomenon would have complicated observing an effect the assumption that RND-type efflux pumps do not act of the pump on toxic substances at intermediate expres- on short-chain alcohols . A more recent study applied sion levels, we first tested whether expression via P BAD random mutagenesis to the AcrAB-TolC efflux pump of is homogenous in the case of P. putida by placing GFP E. coli, leading to the identification of mutations which under control of P and determining fluorescence after BAD allow export of n-butanol, isobutanol, and n-heptanol induction with 0, 1, 10, and 100 mM l-arabinose. Using . The pump was introduced into n-butanol-produc - flow cytometry, we found that GFP fluorescence was ing E. coli and shown to increase titers . Unfortunately, quantitative and homogenous (Additional file 2: Figure this approach for engineering efflux pumps is tedious and S2), demonstrating that the promoter can be used effi - limited by the plasticity of the native pump for broad- ciently with P. putida for fine-tuning of expression levels. ening its substrate specificity with respect to the target To determine the burden of TtgABC pump expression compound. Hence, the discovery of efflux pumps acting in P. putida, we grew the plasmid-carrying strain with natively on a desired target product is a promising alter- different inducer concentrations. Growth was unaffected native approach for improving the tolerance of microbial when using up to 50 mM of l-arabinose (Fig. 1a). Induc- production strains to support metabolic engineering for tion with very high levels of l-arabinose (100 mM) led to biofuel production. a prolonged lag-phase of about 2 h, but the strain then We show that controlled expression of the TtgABC fully recovered growth. For comparison, we expressed efflux system led to a strong increase in survival rate of TtgABC in an E. coli strain engineered to give homoge- P. putida DOT-T1E exposed to high concentrations of nous expression using P . We found that TtgABC BAD n-butanol, isobutanol, isoprenol, and isopentanol. We expression in this strain inhibits growth starting from observed increased extracellular n-butanol concentra- l-arabinose concentrations as low as 1 μM (Fig. 1b), indi- tions when cells expressing the efflux pump were incu - cating that transfer of the pump to other strains may be bated in n-butanol containing media, indicating active challenging due to toxicity (see “Discussion”). efflux. This suggests that the TtgABC efflux pump can be We next determined the inducer concentrations, used to improve production of a number of short-chain induction times, and growth stage which maximize the alcohols. On the other hand, we found that growth rates effect of TtgABC expression on survival in high concen - in n-butanol and isopentanol were not increased when trations of n-butanol (“Methods”). Although previous expressing the pump, hinting at the limitations of growth results indicated that P. putida DOT-T1E could grow in assays for the identification of efflux pumps with novel the presence of up to 6% n-butanol (v/v) after long-term functions. To our knowledge, this is the first report of adaptation , these results could not be reproduced a native RND-type efflux pump shown to export short- by us or others . Instead, we observed that the wild- chain alcohols. type strain could survive up to 1.9% of n-butanol for 2 h, and that its tolerance as measured by growth or sur- Results vival did not increase even after 2 months of adaptation Expression of efflux pumps can be toxic and their oper - (“Methods”). ation requires energy , which implies a tradeoff The TtgABC plasmid-carrying strain was then grown between the benefit and burden of pump expression in with and without induction, and the plasmid-free wild- toxic environments . Therefore, the function of an type strain was grown with l -arabinose as additional efflux pump critically depends on the level of expres - control. Induction of TtgABC expression with 1.5 mM sion, which must be fine-tuned to maximize its effect l -arabinose and growth to stationary phase resulted in the while minimizing the burden. To gain quantitative con- largest increase of survival rate, and these conditions were trol of expression and fine-tune expression levels to chosen for all subsequent experiments. Note that TtgABC avoid a negative effect on cell survival from overexpres - expression did not affect growth up to 50 mM l -arabinose sion, we placed the TtgABC operon under control of the (cf. Fig. 1a). After normalizing the cell densities, cells were Basler et al. Biotechnol Biofuels (2018) 11:136 Page 3 of 10 To assess whether the observed increase in survival was indeed due to efflux of the alcohol, rather than a side effect or general stress response due to expressing the membrane protein, we compared extracellular n-butanol concentrations of non-induced and induced cells after incubation for 110 min with 0.2% n-butanol via gas chro- matography with a flame ionization detector (GC-FID, “Methods”). The rationale for this approach is that strains with higher efflux accumulate less intracellular n-butanol, leading to higher measurable extracellular concentra- tions. The approach was previously used to measure n-butanol efflux . We found a significant increase of 5.8% (t test, p value < 0.05) in extracellular n-butanol con- centrations of induced cells, which corresponds to about −8 −1 −1 2.9 × 10 mol gDW s more efflux of n-butanol from induced cells compared to non-induced cells. We deter- mined the viable cell numbers before and after incuba- tion to ensure that cells were not dying, since this could have led to a release of the alcohol due to membrane disintegration (“Methods”). Hence, our results indicate active efflux of n-butanol in cells expressing TtgABC. The inner membrane pump of TtgABC is TtgB, which is the component responsible for substrate binding. TtgB is homologous to AcrB in the AcrAB-TolC efflux pump of E. coli (65% amino acid sequence similarity). Three individual point mutations of AcrB (M355L, I466, S880P) were previously shown to allow export of n-butanol, isob- utanol, and n-heptanol in E. coli . Sequence align- ment revealed that the same amino acids are present at the corresponding loci of TtgB, with strongly conserved regions surrounding M355 and I466. To test whether increased efflux of n-butanol could be achieved in Fig. 1 Toxicity of TtgABC expression. The pBbB8k‑ TtgABC plasmid TtgABC, we applied each of the three point mutations, as was introduced into P. putida DOT‑ T1E and E. coli DP10, and efflux well as the combination of the three, at the corresponding pump expression was induced using different concentrations of l ‑arabinose. Growth curves indicate a prolonged lag‑phase when loci of TtgB (“Methods”). We found that each of the three inducing expression with 100 mM l ‑arabinose in P. putida (a), mutations, as well as their combination, resulted in a while growth of E. coli is strongly affected starting from inducer decrease of survival rate compared to the wild-type strain concentrations of 1 µM (b) (not shown). This may indicate structural differences of the inner membrane pumps, despite their homology, or a generally higher specificity of TtgB to n-butanol, com - incubated with 1.9% n-butanol, 2.2% isobutanol, 1.5% pared to AcrB, preventing a further increase of n-butanol isoprenol, or 0.7% isopentanol for 2 h, which represent efflux in TtgABC. slightly sub-lethal concentrations. Survival rates were cal- To assess whether the effect of efflux pump expression culated by sampling the number of living cells before and on cell survival implies faster growth in biofuels, we grew after incubation (Fig. 2, “Methods”). For isobutanol and the induced and non-induced plasmid-carrying strains, isoprenol, survival of the wild-type was at least an order as well as the plasmid-free wild-type strain, in a plate of magnitude higher compared to the plasmid-carrying, reader with a range of n-butanol and isopentanol concen- non-induced variants. This indicates that the presence of trations (Figs. 3, 4). Growth rates were largely unaffected the plasmid may decrease survival, which could limit the up to concentrations of 0.125% (v/v), and the wild-type observable effect in strains expressing the pump. For each −1 strain grew slightly faster (0.26–0.35 h ) compared of the four alcohols tested, survival of the cells expressing −1 to non-induced (0.19–0.34 h ) and induced (0.18– TtgABC was at least tenfold higher compared to the non- −1 0.31 h ) plasmid-carrying strains in both alcohols. The induced and wild-type strains. effect was even more pronounced at 0.25% (v/v), where Basler et al. Biotechnol Biofuels (2018) 11:136 Page 4 of 10 Fig. 2 TtgABC expression increases survival of Pseudomonas putida DOT‑ T1E in biofuels. Fraction of P. putida DOT‑ T1E cells surviving 2 h of incubation with a 1.9% n‑butanol b 2.2% isobutanol, c 1.5% isoprenol, and d 0.7% isopentanol. Cells carrying TtgABC on a plasmid with l ‑arabinose + − inducible promoter were grown with (ttgABC ) or without (ttgABC ) inducer. Plasmid‑free wild‑type cells ( WT ) were grown with l ‑arabinose as control. Ratios were obtained by sampling viable cell numbers before and after incubation. The error bars indicate minimum and maximum of 3–6 biological replicates −1 growth of all strains was slightly affected, and the wild- to the induced strain (0.26 h ). Hence, the observed −1 type strain grew faster (0.23–0.28 h ) compared to the effect of TtgABC on survival in highly toxic conditions −1 −1 non-induced (0.14–0.21 h ) and induced (0.11–0.2 h ) does not translate into increased growth in less toxic bio- plasmid-carrying strains. Moreover, there was no differ - fuel concentrations. Accordingly, growth assays could not ence in growth rates between non-induced and induced have identified the observed effect of the TtgABC efflux strains for any of the tested concentrations. To account pump on short-chain alcohols. for the possibility of oxygen limitation in plate readers due to the small volumes used, we also tested growth in Methods larger volumes (Additional file 3: Figure S3, “Methods”). Plasmid construction and survival assays Without alcohol, growth rates were higher compared to We amplified ttgABC (5762 bp) from the genome of P. −1 the plate reader and similar for all strains (0.53 h ). With putida DOT-T1E and cloned it on a broad host range 0.5% n-butanol, growth rates of the wild-type and non- plasmid (BBR1) between the l-arabinose inducible P BAD −1 induced strains were slightly higher (0.29 h ) compared promoter and double terminator using Gibson Assembly. Basler et al. Biotechnol Biofuels (2018) 11:136 Page 5 of 10 Fig. 3 TtgABC expression does not increase growth of P. putida DOT‑ T1E in n‑butanol. Growth of P. putida DOT ‑ T1E without plasmid ( WT ), without − + induction (ttgABC ), and with induction (ttgABC ) in 0% (a), 0.25% (b), and 0.5% (c) n‑butanol. Growth rates of the wild‑type strain are slightly higher, and there is no increase in growth rate due to TtgABC expression (d) Long‑term adaptation of P. putida to n‑butanol The final construct, termed pBbB8k-TtgABC after the Pseudomonas putida DOT-T1E was grown for 2 months Bgl-Brick standard  has a length of 9529 bp and fur- in 20 mL LB medium with 33 mM l-arabinose, 0.5% ther contains araC and Kanamycin resistance marker. For n-butanol, and 50 µg/mL rifampycin. The medium was measuring expression levels, we created another plasmid renewed daily by sub-culturing of 100 µL into fresh containing GFP between the P promoter and double BAD medium. After 1 month, the strain was split into two cul- terminator. The plasmids were transformed into P. putida tures. One culture was continued with 0.5% n-butanol in DOT-T1E and E. coli DP10  using electroporation the medium, while the other was exposed to increasing (Bio-Rad MicroPulser). concentrations of up to 1% n-butanol, with renewal of For the survival assays, three to six colonies were inoc- medium after reaching a visible density (approximately ulated in LB medium with 1.5 mM (induced) and with- every 3 days). After 2 months, the strains were exposed to out (non-induced) l-arabinose and incubated shaking at a range of n-butanol concentrations to test for increased 200 rpm and 30 °C overnight. After reaching stationary tolerance. No difference was observed in survival or phase, all samples were diluted to an equal optical density growth between the strains undergoing the long-term of 0.1 at 600 nm, and 2 mL of diluted culture (approxi- adaptation and the original wild-type strains. mately 1.6 × 10 cells) was incubated shaking at 200 rpm and 30 °C in an airtight 15-mL plastic tube with the cor- Quantification of extracellular n‑butanol using GC‑FID responding alcohol for 2 h. Diluted cell cultures were Three biological replicates were grown overnight with plated on selective medium (plasmid-carrying variants) and without 1.5 mM l -arabinose as described in “Plas- and LB (plasmid-free WT) before and after incubation, mid construction and survival assays”. Cell numbers were and colonies were counted to determine the percent of equalized by determining optical density and transferring surviving cells. Basler et al. Biotechnol Biofuels (2018) 11:136 Page 6 of 10 Fig. 4 TtgABC expression does not increase growth of P. putida DOT‑ T1E in isopentanol. Growth of P. putida DOT‑ T1E without plasmid ( WT ), without − + induction (ttgABC ), and with induction (ttgABC ) in 0% (a), 0.25% (b), and 0.5% (c) isopentanol. Growth rates of the wild‑type strain are slightly higher, and there is no increase in growth rate due to TtgABC expression (d) an adjusted volume of ~ 50 mL at OD600 of 1.38 into as internal standard and quantified using authentic stand - centrifuge falcons. Assuming 1 OD600 unit corresponds ards. The measurements of the three replicates of induced 8 −1 to 8 × 10 cells mL , this density corresponds to about and non-induced strains are shown in Additional file 4 . 5.5 × 10 cells. Samples were split in half and cultures To test whether cells were dying during incubation, were centrifuged three times at 4000×g and washed with diluted cell cultures were plated before and after incuba- PBS at pH = 6.5. The last pellet was re-suspended in PBS tion with 0.2 and 1% n-butanol, and the resulting colo- with 0.2% n-butanol to reach a final weight of 1 g, roughly nies counted. Although we observed a slight decrease in half of which were cells. Cells were incubated at room cell numbers, this is likely due to a loss of sample when temperature for 110 min, and then centrifuged at 4000×g. extracting the supernatant, because cell counts were sim- 250 µL of supernatant was added to 250 µL of ethyl acetate ilar at 0.2 and 1% n-butanol (Additional file 4). Moreover, with isoprenol as internal standard, vigorously vortexed there was no significant difference in cell counts between for 15 min, and centrifuged at 22×g for 1 min. 100 µL of induced and non-induced cells after incubation (t test, the organic phase was then removed for further analysis. p value 0.26), indicating that pump expression does not Alcohols were quantified using a Tr-Wax column lead to cell death under these conditions. (0.25 mm by 30 m, 0.25-μm film thickness; Thermo Elec - tron) on a Focus GC apparatus with a TriPlus autosampler Transfer of point mutations from AcrB to TtgB (Thermo Electron). Hydrogen was used as a carrier, and Three mutations of AcrB were previously shown to indi - was set at a constant pressure of 300 kPa, with the inlet vidually increase efflux of n-butanol, isobutanol, and temperature set to 200 °C. The oven program was 40 °C n-heptanol . We applied the three point mutations for 1.5 min and then it was increased from 40 to 110 °C M355L, I466F, and S878P (corresponding to S880P in −1 at 15 °C min . Samples were normalized using isoprenol the aligned AcrB) individually, and the combination of Basler et al. Biotechnol Biofuels (2018) 11:136 Page 7 of 10 the three, to TtgB by PCR amplification with mismatch For example, knockout of AcrAB in E. coli did not affect primers and Gibson Assembly of the resulting ampli- cell survival in high concentrations of ethanol or 1-pro- cons. The created constructs pBbB8k-TtgABC[M355L], panol compared to the wild-type strain, regardless of pBbB8k-TtgABC[I466F], pBbB8k-TtgABC[S878P], and whether or not the pump was induced with the mar phe- pBbB8k-TtgABC[M355L,I466F,S878P] were sequenced notype . The authors concluded that the AcrAB-TolC and transformed into P. putida DOT-T1E via electropo- efflux pump does not act on short-chain alcohols. Simi - ration. The strains were then subjected to the survival larly, although mutations in AcrA increased tolerance assay for n-butanol described in “Plasmid construction of E. coli to isobutanol, measured by optical density of and survival assays”. the cultures after treatment with the alcohol for 24 h, its knockout reduced tolerance . This suggests that AcrA Growth curves does not have a (positive) effect on isobutanol tolerance Growth rates were estimated via a microplate reader with respect to growth. kinetic assay. Overnight cultures of P. putida grown in LB Moreover, in a large-scale screening approach , medium at 30 °C were diluted 1:100 into fresh LB media 43 efflux pumps, including TtgABC and AcrAB-TolC, with either 1.5 mM (induced) or 0 mM (non-induced) were tested for their effect on seven relevant biofuels. l-arabinose and various concentrations of twofold All pumps were cloned into E. coli, and the resulting diluted n-butanol or isopentanol in 96-well plates (Fal- library of strains was challenged with the biofuels result- con, 353072). Plates were sealed with a gas-permeable ing in competition for growth. Even though several efflux microplate adhesive film (VWR, USA), and then opti - pumps acting on other biofuels were identified by the cal density and fluorescence were monitored for 22 h in study, no pump was found to act on n-butanol or iso- an Infinite F200 Pro (Tecan Life Sciences, San Jose, CA) pentanol. Specifically, TtgABC was found not to have plate reader at 30 °C. Optical density was measured at an effect on these compounds. However, the study dif - 590 nm every 15 min. In between reads, the plate was fers from the approach presented here in two important shaken at a linear amplitude of 6 mm. All data were ana- aspects: first, E. coli was used as a host to express the het - lyzed using custom Python scripts. Growth rates were erologous pumps, and second, growth, rather than sur- calculated via a 10-timepoint sliding window on optical vival, was used to assess biofuel toxicity. We found that densities normalized to a 1-cm path length, where the expression of TtgABC in E. coli inhibits growth (Fig. 1b), maximal slope with an r > 0.95 was reported as the maxi- indicating that toxicity of the heterologous pump could mal growth rate. Growth curves were also done in 15 mL have masked its effect on biofuel tolerance. Moreover, we conical tubes containing 3 mL medium (Additional file 3: also did not observe any effect of the pump on growth Figure S3), incubated shaking at 200 rpm at 30 °C. Optical of P. putida in medium supplemented with low concen- densities for this experiment were measured at 600 nm. trations of biofuels (cf. Fig. 3). Hence, it is likely that the function of the pump is to improve survival under highly Discussion toxic conditions, rather than enabling faster growth. Short-chain alcohols are of great interest as biofuels Consequently, a competitive growth assay could not have because of their high energy–density, superior transport identified the effect of the TtgABC pump on biofuels. capabilities, and possibility of directly replacing common We expect that the production and recovery of the engine fuels as drop-ins. Several engineering strategies short-chain alcohols n-butanol, isobutanol, isopentanol demonstrated the successful production of short-chain and isoprenol can be directly improved by expressing alcohols, including n-butanol, isobutanol, isopentanol, TtgABC in the production strain. We only tested these and isoprenol using E. coli [28–31]. However, their pro- four short-chain alcohols here, and found that TtgABC duction is strongly affected by the toxicity of the product, acts on each of them. It is therefore likely that the pump leading to suboptimal yields. Several examples demon- also acts on other short-chain alcohols, many of which are strate that the use of efflux pumps can confer increased promising biofuel candidates: for example, the structur- product yields [4–7]. For example, expression of the ATP- ally similar 1-propanol, isopropanol, 1-pentanol, prenol, binding cassette transporter MdlB of E. coli increased and 2-methyl-1-butanol have been successfully produced tolerance to and production of isopentenol, although no with engineered bacteria [28, 33–35]. To our knowledge, effect on other short-chain alcohols was observed . no efflux pump is known yet to act on these compounds, This suggests that MdlB acts on isopentenol, even though and TtgABC would be a promising candidate to test for efflux was not shown directly. decreasing toxicity and improving their production. It was considered unlikely that efflux pumps of the The observed toxicity of expressing the pump in E. coli type RND act on short-chain alcohols, because all pumps (cf. Fig. 1b) indicates that directly transferring TtgABC tested so far did not have an effect on these biofuels . to other strains may be challenging. Recently identified Basler et al. Biotechnol Biofuels (2018) 11:136 Page 8 of 10 mutations in E. coli leading to a reduced burden of express- expression of the efflux pump and identification of the con - ing membrane proteins may help to reduce the toxicity of ditions which maximize its effect on tolerance open the pos - the efflux pump , as it may chaperone overexpression sibility of directly testing the effect of the pump on other . Strategies for regulating membrane protein produc- toxic substances, further engineering its substrate specific - tion via genetic circuits are also promising . Moreover, ity, and its use in metabolic engineering applications. P. putida may itself serve as production host, and several Moreover, we demonstrated that the specific condi - studies have already demonstrated that the strain can be tions and levels of pump expression, as well as the type engineered as a robust biocatalyst. Plenty of genetic tools of assay used to determine toxicity, can have a great influ - are available for modifying the genetic potential of the ence on the observed effect of the efflux pump. Hence, strain [38–40]. It has been metabolically engineered for our results indicate that the in-depth study of an indi- the production of a large variety of compounds, includ- vidual efflux pump, in addition to large-scale screen - ing polyketides, non-ribosomal peptides, rhamnolipids, ing approaches, may lead to the identification of efflux and aromatic and non-aromatic compounds (reviewed in pumps with novel functions. The techniques employed ). The strain has also great potential for the direct pro - here are easy to implement and can be readily applied to duction of biofuels from lignocellulosic biomass, as it has other strains and efflux pumps. Consequently, it is possi - been successfully engineered for production of aromatic ble that novel efflux pump functions for use in microbial compounds from lignin components [42, 43]. The strain production of biofuels or other toxic substances will be was also successfully engineered to grow under anaerobic discovered using the presented approach. conditions [44–46], and it was used in two-phase liquid This is the first report of a native RND-type efflux extraction systems [47, 48], facilitating the production and pump acting on short-chain alcohols, an important class extraction of toxic apolar compounds in bioreactors. of next-generation biofuels. Together with the availability More specifically, P. putida was engineered for pro - of genetic tools, catabolic diversity, versatile metabolism duction of n-butanol by introducing the biosynthetic and high intrinsic tolerance of the strain to various toxic pathway from C. acetobutylicum . Since P. putida substances , we believe that P. putida will become degrades n-butanol , it will be important to knock an important biocatalyst for biofuels. To this end, the out the corresponding genes to further increase yields TtgABC efflux pump will provide a valuable tool for . Hence, the next step will be to express TtgABC in reducing toxicity and increasing yields. a production strain deficient of n-butanol consumption Additional files and determine the increase in yield. Since we observed a potential burden of the introduced plasmid on survival Additional file 1: Figure S1. Plasmid stability of pBbB8k ‑ TtgABC in and growth (cf. Figs. 2b, c), an alternative approach would Pseudomonas putida DOT‑ T1E. The plasmid‑ carrying strain was grown be to modify the regulatory region of TtgABC on the for 5 consecutive days with 2 mM L‑arabinose for inducing expression of TtgABC and without kanamycin. The medium was renewed once per genome to obtain control over efflux pump expression day. Plasmid stability was determined by plating on kanamycin plates and without the need of a plasmid, e.g., using classical trans- comparison of viable cell numbers. poson-based techniques . Although a recent study Additional file 2: Figure S2. Cell‑level GFP expression using the L ‑arab ‑ indicates that genome editing using lambda Red based inose inducible P promoter in Pseudomonas putida DOT‑ T1E. Cells car‑ BAD recombineering of P. putida KT2440 is possible, our rying the pBbB8k‑ GFP plasmid were induced overnight without (brown), with 1 mM (red), 10 mM (purple), and 100 mM (green) L‑arabinose. attempts with the strain DOT-T1E were not successful. Cell‑level fluorescence of GFP was measured using flow cytometry. The An alternative approach would be to identify plasmids distributions indicate a homologous and quantitative increase of expres‑ which do not affect the strain when exposed to toxic sub - sion at the cell level. stances, such as plasmids shown to represent a minimal Additional file 3: Figure S3. TtgABC expression does not increase growth of P. putida DOT‑ T1E in n‑butanol. Growth of P. putida DOT ‑ T1E without burden for growth in several strains of Pseudomonas . - + plasmid ( WT ), without induction (ttgABC ), and with induction (ttgABC ) With the help of such approaches, the effect of TtgABC in 0% (a) and 0.5% (b) n‑butanol in 15 mL conical tubes containing 3 mL expression on tolerance to short-chain alcohols can be medium. Similar to the results obtained from the plate reader (cf. Fig. 3), growth of the wild‑type strain is slightly faster with and without n‑butanol, further optimized with the aim of maximizing yields in and TtgABC expression does not increase growth in n‑butanol. biofuel production strains. Additional file 4. Quantification of extracellular n‑butanol using GC‑FID. The spreadsheet contains the estimated fraction of viable cells after Conclusions incubation in PBS containing 0.2 and 1% n‑butanol, and the measure ‑ ment values of quantified extracellular n‑butanol in induced (ttgABC ) and We showed that the TtgABC efflux pump in P. putida non‑induced strains (ttgABC ). The original readings for n‑butanol and the DOT-T1E improved tolerance to four short-chain alcohols internal standard isoprenol are shown, as well as the n‑butanol measure ‑ and increases efflux of n -butanol. It is likely that the efflux ments normalized by the standard, and the derived concentrations as volume per volume and milligrams per milliliter. The inlay shows the pump acts on other yet undetermined compounds of indus- n‑butanol standard curve, indicating that concentrations are in a linear trial relevance, particularly short-chain alcohols. Controlled measurement range. Basler et al. Biotechnol Biofuels (2018) 11:136 Page 9 of 10 Authors’ contributions 4. Dunlop MJ. Engineering microbes for tolerance to next‑ gen‑ GB, DTE, and JK designed the research. GB and MT performed experiments eration biofuels. Biotechnol Biofuels. 2011;21(4):32. https ://doi. and analyzed the data. GB wrote the paper. All authors approved the final org/10.1186/1754‑6834‑4‑32. manuscript. 5. Dunlop MJ, et al. Engineering microbial biofuel tolerance and export using efflux pumps. Mol Syst Biol. 2011;10(7):487. https ://doi.org/10.1038/ Author details msb.2011.21. Department of Chemical and Biomolecular Engineering, University of Cali‑ 6. Boyarskiy S, et al. Transcriptional feedback regulation of efflux protein fornia, Berkeley, CA, USA. Max Planck Institute for Molecular Plant Physiology, expression for increased tolerance to and production of n‑butanol. Metab Potsdam, Germany. Department of Plant & Microbial Biology, University Eng. 2016;33:130–7. https ://doi.org/10.1016/j.ymben .2015.11.005. of California, Berkeley, CA, USA. Joint BioEnergy Institute, Emeryville, CA, 7. Wang J‑F, Xiong Z ‑ Q, Li S‑ Y, Wang Y. Enhancing isoprenoid production USA. Department of Chemical and Biological Engineering, Northwestern Uni‑ through systematically assembling and modulating efflux pumps in versity, Evanston, IL, USA. Chemistry of Life Processes Institute, Northwestern Escherichia coli. Appl Microbiol Biotechnol. 2013;97:8057–67. https ://doi. University, Evanston, IL, USA. Center for Synthetic Biology, Northwestern org/10.1007/s0025 3‑013‑5062‑z. University, Technological Institute B486, Evanston, USA. Biological Systems 8. Ornston LN. Regulation of catabolic pathways in Pseudomonas. Bacteriol and Engineering Division, Lawrence Berkeley National Laboratory, Berkeley, Rev. 1971;35(2):87–116. CA, USA. Novo Nordisk Foundation Center for Sustainability, Technical Univer‑ 9. Worsey MJ, Williams PA. Metabolism of toluene and xylenes by Pseu- sity of Denmark, Copenhagen, Denmark. domonas (putida (arvilla) mt‑2: evidence for a new function of the TOL plasmid. J Bacteriol. 1975;124(1):7–13. Acknowledgements 10. Inoue A, Horikoshi K. A Pseudomonas thrives in high concentrations of We thank Sergey Boyarskiy for support with creating the plasmids, Jose Anto‑ toluene. Nature. 1989;338:264–6. https ://doi.org/10.1038/33826 4a0. nio Reyes Darias for help with the adaptation experiments, and Jessica Trinh 11. Kieboom J, et al. Identification and molecular characterization of an for performing the flow cytometry experiments. efflux pump involved in Pseudomonas putida S12 solvent tolerance. J Biol Chem. 1998;273(1):85–91. Competing interests 12. Rühl J, et al. Selected Pseudomonas putida strains able to grow in the JDK has financial interests in Amyris, Lygos, Constructive Biology, Demetrix, presence of high butanol concentrations. Appl Environ Microbiol. and Napigen. 2009;75(13):4653–6. https ://doi.org/10.1128/AEM.00225 ‑09. 13. Cuenca MDS, et al. Understanding butanol tolerance and assimilation Availability of data and materials in Pseudomonas putida BIRD‑1: an integrated omics approach. Microb The datasets used and/or analyzed during the current study are available from Biotechnol. 2016;9(1):100–15. https ://doi.org/10.1111/1751‑7915.12328 . the corresponding author on reasonable request. 14. Ramos JL, et al. Isolation and expansion of the catabolic potential of a Pseudomonas putida strain able to grow in the presence of high concen‑ Consent for publication trations of aromatic hydrocarbons. J Bacteriol. 1995;177(14):3911–6. Not applicable. 15. Ramos JL, et al. Efflux pumps involved in toluene tolerance in Pseu- domonas putida DOT‑ T1E. J Bacteriol. 1998;180(13):3323–9. Ethics approval and consent to participate 16. Nikolouli K, Mossialos D. Functional characterization of TtgABC efflux Not applicable. pump of the RND family in the entomopathogenic bacterium Pseu- domonas entomophila. Annals of Microbiology. 2016;66(1):499–503. https Funding://doi.org/10.1007/s1321 3‑015‑1119‑9. GB was supported by the Max Kade Foundation and the Max Planck Society. 17. Ramos JL, et al. Mechanisms of solvent resistance mediated by inter‑ This work was part of the DOE Joint BioEnergy Institute (http://www.jbei. play of cellular factors in Pseudomonas putida. FEMS Microbiol Rev. org/), which was supported by the U.S. Department of Energy, Office of 2015;39(4):555–66. https ://doi.org/10.1093/femsr e/fuv00 6. Science, Office of Biological and Environmental Research, through contract 18. Rojas A, et al. Three efflux pumps are required to provide efficient DE‑AC02‑05CH11231 between Lawrence Berkeley National Laboratory and tolerance to toluene in Pseudomonas putida DOT‑ T1E. J Bacteriol. the U.S. Department of Energy. The United States Government retains and 2001;183(13):3967–73. the publisher, by accepting the article for publication, acknowledges that 19. Ankarloo J, et al. Escherichia coli mar and acrAB mutants display no toler‑ the United States Government retains a nonexclusive, paid‑up, irrevocable, ance to simple alcohols. Int J Mol Sci. 2010;11(4):1403–12. https ://doi. worldwide license to publish or reproduce the published form of this manu‑org/10.3390/ijms1 10414 03. script, or allow others to do so, for United States Government purposes. The 20. Atsumi S, et al. Evolution, genomic analysis, and reconstruction of isobu‑ Department of Energy will provide public access to these results of federally tanol tolerance in Escherichia coli. Mol Syst Biol. 2010;21(6):449. https :// sponsored research in accordance with the DOE Public Access Plan (https :// doi.org/10.1038/msb.2010.98. energ y.gov/downl oads/doe‑publi c‑acces s‑plan). 21. Fisher MA, et al. Enhancing tolerance to short‑ chain alcohols by engineering the Escherichia coli AcrB efflux pump to secrete the non‑ native substrate n‑butanol. ACS Synth Biol. 2014;3(1):30–40. https ://doi. Publisher’s Note org/10.1021/sb400 065q. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 22. Pos KM. Drug transport mechanism of the AcrB efflux pump. Biochim lished maps and institutional affiliations. Biophys Acta. 2009;1794(5):782–93. https ://doi.org/10.1016/j.bbapa p.2008.12.015. Received: 5 January 2018 Accepted: 28 April 2018 23. Turner WJ, Dunlop MJ. Trade‑ offs in improving biofuel tolerance using combinations of efflux pumps. ACS Synth Biol. 2015;4(10):1056–63. https ://doi.org/10.1021/sb500 307w. 24. Siegele DA, Hu JC. Gene expression from plasmids containing the araBAD promoter at subsaturating inducer concentrations represents mixed References populations. Proc Natl Acad Sci USA. 1997;94(15):8168–72. 1. Isken S, de Bont JA. Bacteria tolerant to organic solvents. Extremophiles. 25. Pitera DJ, et al. Balancing a heterologous mevalonate pathway for 1998;2(3):229–38. improved isoprenoid production in Escherichia coli. Metab Eng. 2. Nikaido H. Multidrug efflux pumps of gram‑negative bacteria. J Bacteriol. 2007;9(2):193–207. https ://doi.org/10.1016/j.ymben .2006.11.002. 1996;178(20):5853–9. 26. Halan B, et al. Growth of Pseudomonas taiwanensis VLB120∆C biofilms in 3. Segura A, et al. Solvent tolerance in gram‑negative bacteria. Curr the presence of n‑butanol. Microb Biotechnol. 2017;10(4):745–55. https :// Opin Biotechnol. 2012;23(3):415–21. https ://doi.org/10.1016/j.copbi doi.org/10.1111/1751‑7915.12413 . o.2011.11.015. Basler et al. Biotechnol Biofuels (2018) 11:136 Page 10 of 10 27. Lee TS, et al. BglBrick vectors and datasheets: a synthetic biology 41. Loeschcke A, Thies S. Pseudomonas putida‑a versatile host for the produc‑ platform for gene expression. J Biol Eng. 2011;20(5):12. https ://doi. tion of natural products. Appl Microbiol Biotechnol. 2015;99(15):6197– org/10.1186/1754‑1611‑5‑12. 214. https ://doi.org/10.1007/s0025 3‑015‑6745‑4. 28. Atsumi S, et al. Non‑fermentative pathways for synthesis of branched‑ 42. Johnson CW, Beckham GT. Aromatic catabolic pathway selection for chain higher alcohols as biofuels. Nature. 2008;451(7174):86–9. https :// optimal production of pyruvate and lactate from lignin. Metab Eng. doi.org/10.1038/natur e0645 0. 2015;28:240–7. https ://doi.org/10.1016/j.ymben .2015.01.005. 29. Zheng Y, et al. Metabolic engineering of Escherichia coli for high‑specific‑ 43. Meijnen JP, et al. Sustainable production of fine chemicals by the solvent ‑ ity production of isoprenol and prenol as next generation of biofuels. Bio‑ tolerant Pseudomonas putida S12 using lignocellulosic feedstock. Int technol Biofuels. 2013;24(6):57. https ://doi.org/10.1186/1754‑6834‑6‑57. Sugar J. 2011;113(1345):24–30. 30. Peralta‑ Yahya PP, et al. Microbial engineering for the production 44. Sohn SB, et al. In silico genome‑scale metabolic analysis of Pseudomonas of advanced biofuels. Nature. 2012;488(7411):320–8. https ://doi. putida KT2440 for polyhydroxyalkanoate synthesis, degradation of aro‑ org/10.1038/natur e1147 8. matics and anaerobic survival. Biotechnol J. 2010;5(7):739–50. https ://doi. 31. Zhang K, et al. Expanding metabolism for biosynthesis of nonnatural org/10.1002/biot.20100 0124. alcohols. Proc Natl Acad Sci USA. 2008;105(52):20653–8. https ://doi. 45. Nikel PI, de Lorenzo V. Engineering an anaerobic metabolic regime in org/10.1073/pnas.08071 57106 . Pseudomonas putida KT2440 for the anoxic biodegradation of 1,3‑ dichlo‑ 32. Foo JL, et al. Improving microbial biogasoline production in Escherichia roprop‑1‑ ene. Metab Eng. 2013;15:98–112. https ://doi.org/10.1016/j. coli using tolerance engineering. MBio. 2014;5(6):e01932. https ://doi.ymben .2012.09.006. org/10.1128/mBio.01932 ‑14. 46. Schmitz S, et al. Engineering mediator‑based electroactivity in the 33. Hanai T, et al. Engineered synthetic pathway for isopropanol production obligate aerobic bacterium Pseudomonas putida KT2440. Front Microbiol. in Escherichia coli. Appl Environ Microbiol. 2007;73(24):7814–8. https ://doi. 2015;10(6):284. https ://doi.org/10.3389/fmicb .2015.00284 . org/10.1128/AEM.01140 ‑07. 47. Neumann G, et al. Energetics and surface properties of Pseudomonas 34. George KW, et al. Metabolic engineering for the high‑ yield production of putida DOT‑ T1E in a two‑phase fermentation system with 1‑ decanol as isoprenoid‑based C5 alcohols in E. coli. Sci Rep. 2015;5:11128. https ://doi. second phase. Appl Environ Microbiol. 2006;72(6):4232–8. https ://doi. org/10.1038/srep1 1128.org/10.1128/AEM.02904 ‑05. 35. Chou HH, Keasling JD. Synthetic pathway for production of five ‑ carbon 48. Blank LM, et al. Metabolic response of Pseudomonas putida during alcohols from isopentenyl diphosphate. Appl Environ Microbiol. redox biocatalysis in the presence of a second octanol phase. FEBS J. 2012;78(22):7849–55. https ://doi.org/10.1128/AEM.01175 ‑12. 2008;275(20):5173–90. https ://doi.org/10.1111/j.1742‑4658.2008.06648 .x. 36. Jensen HM, et al. Improving membrane protein expression and function 49. Nielsen DR, et al. Engineering alternative butanol production platforms using genomic edits. Sci Rep. 2017;7(1):13030. https ://doi.org/10.1038/ in heterologous bacteria. Metab Eng. 2009;11(4–5):262–73. https ://doi. s4159 8‑017‑12901 ‑7.org/10.1016/j.ymben .2009.05.003. 37. Gialama D, et al. Development of Escherichia coli strains that withstand 50. Cuenca MDS, et al. A Pseudomonas putida double mutant deficient in membrane protein‑induced toxicity and achieve high‑level recombinant butanol assimilation: a promising step for engineering a biological bio‑ membrane protein production. ACS Synth Biol. 2017;6(2):284–300. https fuel production platform. FEMS Microbiol Lett. 2016;363(5):fnw018. https ://doi.org/10.1021/acssy nbio.6b001 74.://doi.org/10.1093/femsl e/fnw01 8. 38. Martínez‑ García E, de Lorenzo V. Engineering multiple genomic deletions 51. Bloemberg, et al. Green fluorescent protein as a marker for Pseudomonas in gram‑negative bacteria: analysis of the multi‑resistant antibiotic profile spp. Appl Environ Microbiol. 1997;63(11):4543–51. of Pseudomonas putida KT2440. Environ Microbiol. 2011;13(10):2702–16. 52. Udaondo Z, et al. Analysis of solvent tolerance in Pseudomonas https ://doi.org/10.1111/j.1462‑2920.2011.02538 .x. putida DOT‑ T1E based on its genome sequence and a collection of 39. Martínez‑ García E, de Lorenzo V. Transposon‑based and plasmid‑based mutants. FEBS Lett. 2012;586(18):2932–8. https ://doi.org/10.1016/j.febsl genetic tools for editing genomes of gram‑negative bacteria. Meth‑ et.2012.07.031. ods Mol Biol. 2012;813:267–83. https ://doi.org/10.1007/978‑1‑61779 ‑412‑4_16. 40. de Lorenzo V, et al. Mini‑ Tn5 transposon derivatives for insertion mutagenesis, promoter probing, and chromosomal insertion of cloned DNA in gram‑negative eubacteria. J Bacteriol. 1990;172(11):6568–72. Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your ﬁeld rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions
Biotechnology for Biofuels – Springer Journals
Published: May 11, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera